Papers for

market analysts

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Long-context language models struggle with tracking and pattern learning

LEAPS: Long-context Evaluation of Adaptive Pattern Learning and State Tracking

Abstract: Long-context language models (LCLMs) power agents that act over long trajectories, from improving a codebase to running research experiments to forecasters that condition on lengthy histories. These applications draw on three distinct long-context capabilities: recall, where the answer is a fact present in the context; state tracking, where the outcome can be computed by performing a series of transformations to a specified state; and pattern learning, where the goal is to make a prediction by sampling from a distribution inferred from many observations. Existing benchmarks primarily focus on the first capability. In this work, we introduce LEAPS (Long-context Evaluation of Adaptive Pattern learning and State tracking), a collection of 12 novel datasets, spanning agentic workflows, behavioral modeling, and forecasting of market and world events. By construction, LEAPS is resistant to data contamination, extends to millions of input tokens, and uses reliable metrics. We evaluate 30 API and open models from 8K to 512K input tokens. LEAPS distinguishes models that perform near perfectly on recall tasks, finding that such models' state tracking and pattern learning abilities can vary significantly and degrade much more dramatically with increasing input lengths--even for pattern learning, where more observations should make learning a theoretically easier task. Moreover, thinking models often underperform their base counterparts, exhausting their generation budget or oversimplifying the task. We hope LEAPS can guide the development of the next generation of LCLMs.

Mon 19 OctComputation and Language
The gist
Language models that handle very long inputs are good at remembering facts but have trouble with more complex tasks like tracking changes over time or recognizing patterns from many examples. The authors created new tests called LEAPS to better measure these struggles by using very long texts and tasks beyond just recalling facts. They found many models get worse as inputs get longer, especially on tasks needing more thinking, and sometimes versions designed to think harder don’t perform well. This work helps highlight where future models need improvement for tasks requiring understanding and adapting over long text stretches.
Open → 2610.25919v1

Analytic hierarchy process improves pattern ranking with fewer user queries

Boosting the Learning for Ranking Patterns

Abstract: Pattern mining is a valuable tool for exploratory data analysis, but identifying relevant patterns for a specific user is challenging. Various interestingness mea- sures have been developed to evaluate patterns, but they may not efficiently estimate user-specific functions. Learning user-specific functions by ranking patterns has been proposed, but this requires significant time and training samples. In this paper, we present a solution that formulates the problem of learning pattern ranking functions as a multicriteria decision making problem. Our approach uses Analytic Hierarchy Process (AHP) to elicit weights for different interestingness measures based on user preference. We also propose an active learning mode with a sensitivity-based heuristic to minimize user ranking queries while still providing high-quality results. Experiments show that our approach significantly reduces running time and returns precise pattern ranking while being robust to user error, compared to state-of-the-art approaches.

Mon 12 OctArtificial Intelligence
The gist
Finding interesting patterns that match a user's personal needs is hard because many ways to measure patterns exist but none work well for everyone. The authors propose using a method called Analytic Hierarchy Process to combine different pattern scores based on what the user prefers. They add a smart way to ask the user fewer questions but still learn what they like. Their tests show this method is faster and more accurate than previous ones, even if the user makes some mistakes.
Open → 2610.16511v1

Personal AI assistants lower costs but sellers recoup some gains

Who Keeps the Gains from Personal AI Assistants? Seller Adaptation and the Unassisted in a Language-Model Market Simulation

Abstract: Personal AI assistants are beginning to transact for consumers, and early adopters capture real savings. Whether those savings survive, and what happens to consumers who have no assistant, depends on how sellers respond -- a question single-user evidence cannot answer. We build an agent-based rental market in which language models play consumers, assistants, and six adaptive sellers guided by an algorithmic pricing tool. Half the population receives an assistant under an advisory or an executing mandate; the contract pairs a fee only the renter's physical action avoids with a pre-selected add-on an authorised assistant can cancel online. An analytical benchmark and a behaviourally calibrated rule market supply ex-ante predictions, and paired branches with frozen versus adaptive sellers separate adoption effects from market feedback. Across thirty simulated markets, executing assistants cut adopters' spending by 13.7 USD per renter-day when sellers are frozen; adaptation claws back about a third, leaving 8.7, with the gains arriving both as lower bills and as rentals completed at all. Sellers raise headline rates while cutting fees, and the calibrated forecast of the burden on unassisted consumers (+3.6) does not transfer: their mean spending change is +0.4, confidence interval -0.6 to +1.3. Seller-model swaps and a within-market transfer of fee-setting to the pricing tool show that fee conduct, and with it the division of the gains, is decided on the seller side. Assistants, we conclude, should be evaluated at market level -- completion, total spending, and non-users included -- and the comparison layers locate exactly where a calibrated behavioural forecast fails in a language-model market.

Mon 5 OctMultiagent SystemsArtificial Intelligence
The gist
Personal AI assistants help people save money when renting by negotiating better deals. The authors built a simulated market where AI assistants and sellers interact to see if sellers raise prices to take back some savings. They found that sellers do adapt by increasing headline rates, partially canceling out the savings for users with assistants. People without assistants do not see much difference in their costs. The study suggests that understanding the overall market effects is important when evaluating personal AI assistants.
Open → 2610.05823v1

Logic based model improves human choice predictions with clear rules

Logic-Logit: A Logic-Based Approach to Choice Modeling

Abstract: In this study, we propose a novel rule-based interpretable choice model, Logic-Logit, designed to effectively learn and explain human choices. Choice models have been widely applied across various domains---such as commercial demand forecasting, recommendation systems, and consumer behavior analysis---typically categorized as parametric, nonparametric, or deep network-based. While recent innovations have favored neural network approaches for their computational power, these flexible models often involve large parameter sets and lack interpretability, limiting their effectiveness in contexts where transparency is essential. Previous empirical evidence shows that individuals usually use heuristic decision rules to form their consideration sets, from which they then choose. These rules are often represented as disjunctions of conjunctions (i.e., OR-of-ANDs). These rules-driven, consider-then-choose decision processes enable people to quickly screen numerous alternatives while reducing cognitive and search costs. Motivated by this insight, our approach leverages logic rules to elucidate human choices, providing a fresh perspective on preference modeling. We introduce a unique combination of column generation techniques and the Frank-Wolfe algorithm to facilitate efficient rule extraction for preference modeling---a process recognized as NP-hard. Our empirical evaluation, conducted on both synthetic datasets and real-world data from commercial and healthcare domains, demonstrates that Logic-Logit significantly outperforms baseline models in terms of interpretability and accuracy.

Sun 4 OctMachine Learning
The gist
Understanding how people make choices can be tricky, especially when computer models are complex and hard to explain. The authors created a new kind of model called Logic-Logit that uses simple, understandable rules to mimic how people decide among options. These rules mimic the way people often think in quick, yes-or-no steps to narrow down choices before picking one. Their method makes the models easier to interpret and also more accurate when tested on both made-up and real-world data. This helps in areas where it is important to know not just what decision is made, but also why.
Open → 2610.05501v1

Long lived agents evaluated over weeks of real world replayed data

ReLiveGym: Evaluating Long-Lived Agents over Weeks of Replayed Reality

Abstract: As large language model (LLM) agents become widely adopted, they are increasingly deployed for tasks that require persistent monitoring or recurring actions (e.g., market analysis). These agents are expected to operate unattended for days or weeks, act at the right timing, and adapt to the dynamic environment over time. These challenges are not fully captured in the existing long-horizon agent work, as they often consider a static environment that is not temporally changing. We introduce ReLiveGym, a diagnostic evaluation environment of long-lived tasks in which agents act sparsely over simulated weeks of chronologically replayed real-world news, market, and social-media streams. The tasks span diverse levels of time sensitivity, reasoning intensity, and recurrence. Across eight base language models, we investigate how model choice and harness design affect agent performance on such long-lived tasks. Our results show that how agents determine when to act arises as an important harness-design axis for long-lived tasks; and that the optimal design varies across tasks and sometimes model choices as well. We also evaluate how continuous learning from hindsight feedback affects performance and addresses failure modes observed in these long-lived tasks. These findings indicate model choice, action timing mechanism, and use of feedback as important considerations in the design of long-lived agents. Code: https://github.com/SaharaLabsAI/ReLiveGym

Wed 30 SeptArtificial Intelligence
The gist
Many AI systems are designed to work on tasks that require them to keep watching and acting over several days or weeks, like checking the stock market or news. The authors say that previous tests for these AI agents do not fully capture the challenges of constantly changing real-world environments. They created a tool called ReLiveGym that plays back real news, market, and social media information over simulated weeks, letting AI agents act and learn over a long period. Their experiments show that when and how these agents decide to act is important, and different tasks and AI models may need different designs. They also find that letting agents learn continuously from feedback can help them perform better in these long time tasks.
Open → 2610.00710v1

Polynomial-time method tests peak-oriented preference rationality

A Polynomial-Time Test for Peak-Oriented Rationalizability

Abstract: We study the computational complexity of peak-oriented rationalizability, a survey based revealed-preference test introduced by Seror (2026). We provide a polynomial-time algorithm for testing rationalizability and recovering a utility function, establishing that peak-oriented preference elicitation is computationally tractable. In contrast, we show that computing the peak-oriented Houtman-Maks index is NP-hard. These results delineate the precise computational boundaries of peak-oriented revealed-preference analysis.

Thu 24 SeptComputational Complexity
The gist
Understanding how people choose things can be studied by looking at their most preferred options, or peaks. The authors found a way to quickly check if such peak preferences make sense under a certain theory, and to figure out the ratings behind them. However, another measure related to these preferences is much harder to compute. Their work clarifies exactly which parts of this analysis can be done easily and which cannot.
Open → 2609.30042v1

Calibrating social simulators against real market network data

Testing, not presuming, adequacy: calibrating generative social simulators against emergent network structure

Abstract: Validation of generative social simulators often stops at face validity: emergent network structure is compared descriptively, without quantified parameter uncertainty or an adequacy check. We present an adequacy-aware calibration protocol that couples amortized posterior estimation with a synthetic identifiability assessment, a matched-sample-size adequacy check (prior-predictive reachability plus per-statistic posterior-predictive localization), a diagnosis-guided repair, and a statistic-held-out audit. We demonstrate it on a real second-hand luxury resale market with four channel-by-residency cells, each a bipartite buyer-brand network, using a forward model built from persona profiles elicited once, offline, by a language model. The behavioural parameters are recoverable in all four cells, though calibration is approximate and overconfident for one parameter. The observed summary falls outside the simulator's reachability reference in every cell, with the mean purchased tier as the pervasive discrepancy. The repair meets the value-block criterion in two of four cells but does not restore adequacy, and the held-out audit surfaces a buyer-breadth-dispersion miss no earlier diagnostic detected. A profile-source ablation finds the language-model profiles beat a flat rule baseline in all four cells, yet within-category brand relabelling causes no consistent degradation, so the profiles are a partially validated input whose value rests on structure, not brand identity. Making no causal claim, we conclude that an independent-aggregation account, without agent interaction or a buyer-breadth mechanism, cannot jointly reproduce the market's purchased-tier level, head-brand concentration, community structure and buyer-breadth heterogeneity.

Mon 21 SeptArtificial IntelligenceMultiagent SystemsSocial and Information Networks
The gist
Many social simulators that try to mimic real-world social behaviors stop at just looking similar without checking how accurate their parameters are. The authors present a new detailed method that tests and adjusts these simulators using uncertainties and diagnostic checks. They apply this to a second-hand luxury resale market, showing that while some behaviors in the model can be recovered, key parts of the real market aren’t fully captured by simple models without interactions. Their approach finds where the models miss and helps guide improvements.
Open → 2609.24012v1

Game theory method improves utility recovery from noisy equilibrium data

Suboptimality Loss for Inverse Learning from Imperfect Equilibria

Abstract: Many modern systems involve the strategic interaction of multiple agents. In such settings, observed actions typically reflect equilibrium behavior under utilities that are only partially known. Recovering these hidden utilities from data - the central goal of inverse game theory - is key for prediction, counterfactual analysis, and mechanism design. However, existing approaches based on inverse variational inequalities are highly sensitive to noisy and inconsistent equilibrium observations, thus limiting their applicability. In this paper, we resolve this issue by introducing a game-theoretic suboptimality loss that measures the aggregate utility gain players could obtain by unilaterally deviating from an observed strategy profile. First, we show that this loss is convex and admits an efficient decomposition into player-wise best-responses. Second, we show this loss is sandwiched between the predictability loss and the inverse variational inequality loss, making it a tractable surrogate for equilibrium prediction. Third, we develop a mirror descent algorithm to minimize it and demonstrate on a heterogeneous networked Cournot competition that our approach remains accurate under noisy observations and inconsistent equilibrium data while inverse variational inequality methods produce degenerate estimates.

Sat 19 SeptComputer Science and Game Theory
The gist
Many systems involve multiple people or agents making strategic decisions together. Often, we want to figure out their hidden preferences or goals by looking at their actions, but the data can be noisy or inconsistent. The authors propose a new way to measure how far observed actions are from perfect strategies, which is more reliable under imperfect data. They show that their method works better than previous ones when there is noise or conflicting information.
Open → 2609.23200v1